This extensive literature review, published in Energy Informatics, maps the rise of "Environmentally Sustainable Smart Cities" (ESSC). By analyzing three decades of research, the authors demonstrate how the once-separate domains of "smart" (tech-driven) and "sustainable" (eco-driven) urbanism are finally merging through the convergence of AI, IoT, and Big Data.
TL;DR
This paper provides a comprehensive roadmap of how data-driven technologies are being harnessed to meet global sustainability targets in urban environments. It reveals that while the "smart" and "sustainable" city paradigms were once fragmented, they are now rapidly amalgamating into a single, data-intensive model for decarbonization—though not without significant environmental and ethical costs.
What Problem Does This Paper Address?
By 2050, nearly 75% of the world's population will live in cities. Currently, urban areas consume 75% of global primary resources and are major contributors to greenhouse gas emissions.
Historically, the "Smart City" movement was criticized for being too focused on economic growth and corporate tech-pushing, often at the expense of social and environmental goals. Conversely, "Sustainable Cities" often lacked the high-tech tools needed for real-time resource optimization. The authors address the "research gap" in how these two paradigms are merging. They seek to understand the driving factors, thematic evolution, and the technological "synergies" (specifically AIoT—the Artificial Intelligence of Things) that define modern environmentally sustainable urbanism.
Methods
The study employs a dual-methodological framework:
- Bibliometric Analysis: Using VOSviewer, the authors performed term co-occurrence and citation analysis to map the "knowledge structure" of the field.
- Evidence Synthesis: They combined configurative synthesis (interpreting the "big picture" and thematic evolution) with aggregative synthesis (summarizing empirical findings from multiple studies).
The researchers compartmentalized the literature into three distinct sub-periods (1991–2015, 2016–2019, and 2020–2021) to track how research priorities shifted over time.
Datasets
The authors utilized a curated dataset of 2,574 documents retrieved from the Web of Science (WoS) database.
- Initial Pool: 5,885 documents.
- Final Selection: 2,574 after screening for relevance to the intersection of smart city solutions and environmental sustainability.
- Timeframe: 1991 to early 2022.
Key Insights and Ideas
- The Great Amalgamation: The most significant trend is the transition from fragmented models to integrated "sustainable smart cities." The UN’s Sustainable Development Goals (SDGs), particularly SDG 11, acted as a primary catalyst for this merger.
- COVID-19 as an Accelerator: Surprisingly, the pandemic didn't just disrupt research; it accelerated the "digitalization and decarbonization" agendas. It served as a massive, real-world testbed for smart city technologies.
- The AIoT Powerhouse: The convergence of AI, IoT, and Big Data (AIoT) is the engine of ESSC. IoT provides the "senses" (sensors), Big Data provides the "memory," and AI provides the "intelligence" to optimize energy grids, transport, and waste management.
- The "Green AI" Paradox: While AI helps solve environmental problems, the authors highlight that AI itself has a massive carbon footprint due to the energy-intensive nature of data centers and model training.
Key Results
- Growth Explosion: Research output remained scant until 2010, began rising systematically between 2011–2015, and exploded from 2016 onwards.
- Dominant AI Models: The review identifies Artificial Neural Networks (ANN), Support Vector Machines (SVM), Genetic Algorithms (GA), and Linear Regression as the most frequently applied techniques for urban environmental optimization.
- Thematic Clusters: The authors identified four main research clusters:
- Green: Environmental impacts, climate change, and quality of life.
- Red: Data-driven solutions and AI techniques.
- Yellow: Technological infrastructure and platforms.
- Blue: Synergistic urban and institutional aspects.
Limitations and Future Work
The authors acknowledge several critical areas for future exploration:
- The "Grey Literature" Gap: This study focused solely on peer-reviewed journals. Future work should include reports from local governments and international organizations to capture "on-the-ground" implementation.
- Cost-Benefit Quantification: There is a lack of rigorous studies quantifying the net environmental benefit of these technologies when accounting for their own energy consumption and e-waste.
- Ethical and Regulatory Risks: The paper notes that the "autonomization" of city management poses risks regarding privacy, bias, and "democratic decay" that current regulatory frameworks are ill-equipped to handle.
- Methodological Robustness: More work is needed to develop standardized metrics for capturing the positive impacts of data-driven urbanism.
Summary updated